About

Gerald J. Wang is an Assistant Professor of Civil and Environmental Engineering at Carnegie Mellon University, having joined the faculty in Fall 2019. He completed his Ph.D. in Mechanical Engineering and Computation at the Massachusetts Institute of Technology (MIT), where he conducted research under the guidance of Professor Nicolas G. Hadjiconstantinou. His doctoral work focused on the theoretical and computational modeling of fluid transport phenomena under nanoscale confinement. From 2014 to 2018, he was recognized as a U.S. Department of Energy Computational Science Graduate Fellow (CSGF). Following his Ph.D., Professor Wang performed postdoctoral research with the late Professor James W. Swan in Chemical Engineering at MIT, where he studied the role of friction in dense colloidal suspensions. Prior to his graduate studies, he earned an S.M. in Mechanical Engineering from MIT in 2015 and dual B.S. degrees in Mechanical Engineering and in Mathematics & Physics from Yale University in 2013, where he worked with Professors Nicholas T. Ouellette and Sarah M. Demers. Professor Wang's research expertise lies in nanoscale fluid dynamics and computational modeling, with a focus on understanding fluid transport and frictional phenomena at the nanoscale.

Research topics

• Radiology
• Internal medicine
• Medicine
• Surgery

Selected publications

• Mental health after the sirens: moral injury in responders and survivors of disaster, mass casualty, and crisis care

  Trauma Surgery & Acute Care Open · 2026-04-01

  articleOpen access1st authorCorresponding

  Healthcare professionals responding to disasters, mass casualty incidents, pandemics, and sustained high-acuity clinical environments face extraordinary ethical and emotional demands. While burnout and post-traumatic stress disorder (PTSD) have historically framed clinician distress, these constructs incompletely capture the moral and existential harm experienced when clinicians are forced to act in ways that conflict with deeply held professional and ethical values. Moral injury has emerged as a distinct framework to explain this phenomenon. This review synthesizes current understanding of moral injury in disaster, mass casualty, and crisis care settings, distinguishing it from burnout and PTSD, examines prevalence and impact on clinicians and healthcare systems, identifies unique individual, organizational, and situational risk factors, and reviews mitigation and recovery strategies across the disaster continuum. Critical gaps in measurement and research are highlighted, particularly in command-driven healthcare environments. Addressing moral injury requires system-level interventions, ethical transparency, and leadership accountability to sustain the healthcare workforce beyond the immediate crisis.

  PublisherDOI

• Peering into the crystal ball: Excess entropy scaling predicts equilibrium transport coefficients before equilibration

  The Journal of Chemical Physics · 2026-03-12 · 1 citations

  articleSenior author

  In a molecular-dynamics simulation, an equilibrium transport coefficient is to be computed (as the name suggests) in a system that has reached thermodynamic equilibrium. Indeed, the two most common methods for computing equilibrium transport coefficients-using the Einstein-Helfand (EH) relation and using the Green-Kubo (GK) relation-make the formal assumption that a system has reached thermodynamic equilibrium before sampling begins. There is, however, "no free lunch": Equilibration always demands an upfront computational investment. In this work, we study the question of just how much equilibration is needed for each computational method to yield a serviceable estimate for a transport coefficient, using as our case study a simple fluid that is initialized far out of configurational equilibrium. We show that a third method for computing transport coefficients-excess entropy scaling, which makes use of system structure in the form of the radial distribution function-has several statistically beneficial properties as compared to EH and GK, including faster convergence to the long-time-average value of the transport coefficient and lower sample-to-sample variance en route to convergence, which we rationalize from an information-theoretic perspective. Overall, this work points to the significant value that structure-based estimators may bring to any workflow demanding high-throughput calculation of transport coefficients.

  PublisherDOI

• Emergency Department Thoracotomy Survival After Admission to the Intensive Care Unit: A Secondary Analysis of an Eastern Association for the Surgery of Trauma (EAST) Multi-institutional trial on Damage Control Thoracotomy

  SSRN Electronic Journal · 2026-01-01

  preprintOpen access
  PublisherDOI

• Averaging Molecular Dynamics simulations to study the slow-strain rate behavior of metals

  Open MIND · 2026-01-01

  articleOpen accessSenior author
  OA PDFDOI

• Averaging Molecular Dynamics simulations to study the slow-strain rate behavior of metals

  arXiv (Cornell University) · 2026-03-07

  articleOpen accessSenior author

  The application of molecular dynamics (MD) simulations to quasistatic loading is severely limited by the large separation between atomic vibration timescales and experimentally relevant deformation rates. In this work we employ the Practical Time Averaging (PTA) framework to overcome this limitation and enable atomistic simulations of crystalline solids under quasistatic loading conditions. PTA exploits the intrinsic separation of timescales by defining slow variables as time-averaged observables of the fast atomistic dynamics and their evolution on the slow loading timescale, thereby avoiding explicit integration of the fast dynamics. Using this approach, we simulate uniaxial deformation, in both tension and compression, of 4 to 20 nm cubic specimens of face centered cubic aluminum nanocrystals at applied strain rates approaching quasistatic conditions. We define slow variables as the averaged kinetic energy, potential energy, and normal stress in the loading direction, and track their evolution on the slow timescale. The stress-strain curves show yielding close to the theoretical stress for homogeneous nucleation, followed by successive load drops and rises caused by dislocation nucleation, motion, and exit from free surfaces. The "smaller is harder" effect is evident from the stress-strain response and from the variation of yield stress with sample size. Serrations in the response are more pronounced for smaller samples. The effects of applied strain rate and initial temperature are also studied. The method also captures the evolution of intricate dislocation microstructures on the slow timescale by tracking time-averaged atomic positions. The PTA framework enables simulations at strain rates several orders of magnitude lower than those accessible to conventional MD, demonstrating significant speedup in computational time while retaining full atomistic resolution.

  PublisherOA PDF

• Averaging Molecular Dynamics simulations to study the slow-strain rate behavior of metals

  arXiv (Cornell University) · 2026-03-07

  preprintOpen accessSenior author

  The application of molecular dynamics (MD) simulations to quasistatic loading is severely limited by the large separation between atomic vibration timescales and experimentally relevant deformation rates. In this work we employ the Practical Time Averaging (PTA) framework to overcome this limitation and enable atomistic simulations of crystalline solids under quasistatic loading conditions. PTA exploits the intrinsic separation of timescales by defining slow variables as time-averaged observables of the fast atomistic dynamics and their evolution on the slow loading timescale, thereby avoiding explicit integration of the fast dynamics. Using this approach, we simulate uniaxial deformation, in both tension and compression, of 4 to 20 nm cubic specimens of face centered cubic aluminum nanocrystals at applied strain rates approaching quasistatic conditions. We define slow variables as the averaged kinetic energy, potential energy, and normal stress in the loading direction, and track their evolution on the slow timescale. The stress-strain curves show yielding close to the theoretical stress for homogeneous nucleation, followed by successive load drops and rises caused by dislocation nucleation, motion, and exit from free surfaces. The "smaller is harder" effect is evident from the stress-strain response and from the variation of yield stress with sample size. Serrations in the response are more pronounced for smaller samples. The effects of applied strain rate and initial temperature are also studied. The method also captures the evolution of intricate dislocation microstructures on the slow timescale by tracking time-averaged atomic positions. The PTA framework enables simulations at strain rates several orders of magnitude lower than those accessible to conventional MD, demonstrating significant speedup in computational time while retaining full atomistic resolution.

  PublisherDOI

• Theoretical and experimental study on a novel 3D lattice-structured shock-absorption device produced by digital light processing additive manufacturing

  Journal of Vibration and Control · 2025-01-05

  article

  Digital light processing (DLP) additive manufactured technology uses ultraviolet light to solidify and shape liquid polymers, enabling the creation of intricate devices with viscoelastic properties. In this study, a 3D lattice-structured shock-absorption device using DLP on polyurethane is produced. To better characterize the mechanical behaviors of the device, experimental tests for the additive manufactured device samples are performed and the frequency-dependent force-displacement characteristics are specifically explored. An intriguing phenomenon is observed in which the storage modulus of the device samples initially increases to a peak value with a specific load frequency and subsequently decreases as the frequency further increases. This phenomenon cannot be characterized by the conventional Zener model, where the storage modulus monotonically increases with frequency. To better explain this phenomenon, a modified fractional-derivative Zener (FDZ) model is introduced. In this modified FDZ model, the mechanism of frequency-dependent reversible buckling is introduced to explain the increase-decrease trend of storage modulus. Through the comparison between experimental and theoretical results, it is shown that the modified FDZ model effectively captures the increase-decrease trend of the storage modulus in the device samples and exhibits an overall error of less than 8%, which is a substantial improvement over the Zener model.

  PublisherDOI

• Phonon mode resolved anharmonic heat capacity of solids

  Physical review. B./Physical review. B · 2025-02-05 · 9 citations

  article

  We develop and validate a lattice dynamics framework to include anharmonic effects in the calculation of mode-level phonon heat capacities. To capture anharmonicity, the phonons are renormalized using a temperature-dependent effective potential and a proposed approach based on instantaneous normal modes. Ground-truth total heat capacities are obtained from molecular dynamics simulations. For Lennard-Jones argon (Stillinger-Weber silicon), the deviation of the potential energy contribution to the total heat capacity from the harmonic Dulong-Petit law is $\ensuremath{-}12%$ ($+16%$) at the highest studied temperature of 80 K (1300 K). The mode heat capacities from the lattice dynamics calculations are summed and compared with the ground-truth total heat capacity. For all temperatures considered, the instantaneous normal mode approach gives the best prediction for Lennard-Jones argon (within 1.1%), while for Stillinger-Weber silicon the temperature-dependent effective potential is best (within 0.7%). The Lennard-Jones argon mode heat capacities decrease with increasing frequency and are impacted by the effect of anharmonicity on the mode's self energy and its interactions with other modes. In Stillinger-Weber silicon, the acoustic mode heat capacities increase by up to 30% relative to the Dulong-Petit law, with these deviations driven by the interactions between modes. The proposed calculation framework will improve high-temperature thermal conductivity calculations, where the heat capacity is generally assumed to take on the harmonic value from the Dulong-Petit law.

  PublisherDOI

• Robot-Assisted Radical and Partial Cystectomy

  2025-01-01

  book-chapter
  PublisherDOI

• Damage control thoracotomy trends, techniques, and outcomes: An EAST multicenter trial

  The Journal of Trauma: Injury, Infection, and Critical Care · 2025-05-28 · 1 citations

  article

  BACKGROUND: Damage-control thoracotomy (DCT) lacks evidence regarding frequency of use, optimal technique, and outcomes. This Eastern Association for the Surgery of Trauma multicenter trial aimed to examine DCT usage over the last decade, evaluate types of temporary closure, and assess associated outcomes. METHODS: An international retrospective cohort study of thoracotomies from 2008 to 2020 at 25 centers was performed. Patients age 16 years or older undergoing thoracotomy within 24 hours of admission who survived to intensive care unit (ICU) admission were included. Mixed logistic regression was used to assess complications associated with closure type, trends in DCT utilization, and mortality. Competing risk regression model was used to determine trends in ICU-free days for DCT over time. RESULTS: Nine hundred twenty-two thoracotomy operations were performed, of those 402 (44%) were DCT. Most injuries were penetrating (n = 609, 66%) and the most common mechanism was gunshot wound. Damage-control thoracotomy patients were significantly more injured and ill on presentation. Fifty-four percent of DCT began in the emergency department. Most common temporary closure types included skin only (n = 103, 25%), commercial vacuum device (n = 123,30%), and adhesive dressing (n = 129, 32%). Frequent complications following DCT were pneumonia (n = 57, 14%), acute renal failure (n = 53,13%), and sepsis (n = 41, 10%). Mortality rate in the DCT group was 61%, versus 17% for definitive thoracotomy (n < 0.001). Utilization of DCT has increased in a linear fashion during the study period, as well as ICU-free days out of 30 (odds ratio, 1.66; 95% confidence interval, 1.18-2.33); however, mortality has not changed over time (odds ratio, 0.61; 95% confidence interval, 0.22-1.98). After mixed logistic regression, there was no difference in complications based on closure type. CONCLUSION: The use of DCT is increasing over time with improved ICU-free days, but without improved mortality. Mechanism of temporary closure should be determined based on operator's experience and institutional resources. LEVEL OF EVIDENCE: Therapeutic/Care Management; Level III.

  PublisherDOI

Frequent coauthors

• Sarah Christian

  Stanford University

  45 shared

• Katherine A. Flanigan

  Carnegie Mellon University

  42 shared

• J. Moore

  Carnegie Mellon University

  39 shared

• Douglas S. Scherr

  Institute of Clinical and Experimental Medicine

  38 shared

• Fethiye Ozis

  Northern Arizona University

  36 shared

• Sherven Sharma

  34 shared

• Steven M. Dubinett

  30 shared

• David A. Green

  University of British Columbia

  28 shared

Labs

• M5 LabPI

  The M5 Lab focuses on theoretical and computational modeling of fluid transport phenomena under nanoscale confinement.

Education

• B.S., Mechanical Engineering, Mathematics & Physics

  Yale University

  2013

• Other, Mechanical Engineering

  Massachusetts Institute of Technology

  2015

• Ph.D., Mechanical Engineering and Computation

  Massachusetts Institute of Technology

  2019

Awards & honors

• 2024 CMU Teaching Innovation Award for 'Participation Shouto…